MICHIGAN STATE UNIVERSITY
The Farming Systems Research Group at Michigan State University is drawn from
the departments of Agricultural Economics, Agricultural Engineering, Animal
Science, Crop and Soil Science, Food Science and Human Nutrition, Sociology,
Veterinary Medicine, and supported by the International Agriculture Institute of
M.S.U. and the U.S. Agency for International Development through a matching
strengthening grant under the Title XII program.
Farming Systems Research Group
Michigan State University
The Farming Systems Research Group at Michigan State University, supported
by Title XII Strengthening Grant Funds from the U.S. Agency for International
Development, and administered by the Institute of International Agriculture,
has included Dr. Jay Artis, Department of Sociology; Dr. Robert J. Deans,
Department of Animal Science; Dr. Merle Esmay (and Dr. Robert Wilkinson),
Department of Agricultural Engineering; Dr. Eric Crawford, Department of
Agricultural Economics; Dr. Russell Freed, Department of Crop and Soil
Sciences (also representing Horticulture); Dr. Al Pearson, Department of
Food Science and Human Nutrition; Dr. Tjaart Schillhorn Van Veen, Department
of Veterinary Medicine; with Dr. George Axinn, International Studies and
Programs and Agricultural Economics, Chair, and Ms. Beverly Fleisher,
graduate research assistant.
RESEARCH AND AGRICULTURAL
by Eric W. Crawford
Working Paper No. 1
THE MICHIGAN STATE UNIVERSITY FARMING SYSTEMS RESEARCH GROUP
WORKING PAPER SERIES
Farming Systems Research and Agricul-
Farming Systems Position Paper
Livestock Systems and Animal Health
Issues in Farming Systems Research --
an Agronomist's Perspective
Farming Systems Research As It Relates
To The Animal Sciences
Farming Systems Research Position Paper
The Farming Systems Research Approach in
the Agricultural Engineering Field
Issues in Farming Systems Research --
a Multidisciplinary Behavioral Science
Farming Systems Research and
An M.S.U. Approach to Farming Systems
The M.S.U. Farming Systems Research
A Working Bibliography on Farming
Systems Research August, 1981
Social Impact, Economic Change, and
Development -- with illustrations
Tjaart Schillhorn van Veen
Robert J. Deans
Merle L. Esmay
George H. Axinn
Robert H. Wilkinson
Beverly Fleisher and
George H. Axinn
George H. Axinn and
Nancy W. Axinn
WORKING PAPER #1
Farming Systems Research and Agricultural Economics
by Eric W. Crawford*
In its initial form, this paper was intended to generate discussion
within the Michigan State University Farming Systems Research Working Group.
At the outset, we agreed that it would be desirable to establish a common
understanding of what farming systems research (FSR) was all about. We also
hoped to learn something about the perspective, concepts, and methodology of
the different disciplines represented within our group. As we attempted to
define basic terms such as "farming system," not to mention "farming systems
research," it quickly became apparent that reaching this common understanding
across disciplines would not only be more time-consuming but also more impor-
tant than originally expected.
In the discussion papers, we therefore attempted to set forth our prelim-
inary understanding of FSR, how it related to other types of research and
problem-solving activities in our respective disciplines, and how the involve-
ment of other disciplines would contribute to our own work in FSR. The papers
were not an effort to advance the frontier of any particular discipline, but
rather to communicate enough of the perspective of each discipline so that
henceforth we could operate on the same wavelength. The papers--at least this
one--also reflect a personal viewpoint and a necessarily selective picture of
the disciplines represented.
*Assistant Professor of Agricultural Economics, Michigan State University.
Many thanks to other members of the FSR Working Group, and to Warren Vincent,
for helpful comments on earlier drafts of this paper.
What is Farming Systems Research?
In defining farming systems research (FSR), I feel it is desirable to
distinguish between the following two categories:
1. Research on the farming system (RFS), principally characterized by a
holistic or systems view, i.e., a focus on some or all major elements and
interactions of the farm operation, rather than on individual components such
as particular crop or animal production activities. RFS is therefore a broad
category of research. It encompasses but is not limited to disciplinary
research, i.e., research aimed at improving the theory or analytical methods
of a given discipline, provided the work is systems rather than component
oriented. Examples include simulation or econometric studies of the farm
household. In my view, RFS does not necessarily involve either a multidisci-
plinary team approach or contact with the farmer.
2. Farming systems research (FSR), described by Hildebrand (1977),
CGIAR/TAC 1978), and Gilbert, Norman, and Winch (1980), and carried out
largely in the international arena. To me, FSR is a subset of RFS. Its dis-
tinctive features are: (a) implementation by a multidisciplinary team of
scientists; (b) close contact between researchers and farmers; (c) recognition
of farmer goals and the relationship between the human and technical aspects
of the farming system; (d) an orientation toward generation of locally suited
technology for improving the productivity of the farming system; and (e) an
emphasis on field level rather than experiment station research activity.
The term "farming systems research" has been applied to programs which
vary significantly along several dimensions: (1) degree of farmer involve-
ment; (2) scope in terms of how many farming systems components and linkages
are considered; (3) size of the "recommendation domain" of the research, i.e.,
the breadth of the region or farm type for which the research is relevant; (4)
extent of experiment station versus farm level activity; and (5) degree of
multidisciplinary involvement.1 Programs labelled as FSR include those which
focus only on the cropping system or even more narrowly on particular key
crops such as maize, rice, or wheat, and those which are carried out largely
by biological scientists on the experiment station. Disciplinary studies
which are not oriented to technology generation are not commonly labeled as
FSR, yet I think they can be important sources of knowledge about the farming
system. This is why I define the broad category of RFS.
At the urging of international donors, FSR has been recently promoted and
undertaken largely by researchers concerned with small farm agriculture in the
Third World, whether at national or international research centers. However,
the FSR methodology is potentially suited to small and large farms alike, and
to North American and European farms as well as those in the Third World.
Nonetheless, the capacity of FSR to examine complex multi-enterprise farming
systems and to "give a voice" to the farmer is probably more beneficial in
Third World agriculture than in North America and Europe, where farms tend to
be more specialized and where farmers have the resources and education needed
for them to represent their own interests to the research and extension estab-
1For the sake of comparison, the definition of FSR given by Gilbert, Nor-
man, and Winch (1980) may be quoted in part:
Farming systems research views the farm or production unit and the
rural household or consumption unit--which in the case of small farmers
are often synonymous--in a comprehensive manner. FSR also recognizes the
interdependencies and interrelationships between the natural and human
environments. The research process devotes explicit attention to the
goals of the whole farm/rural household and the constraints on the
achievement of these goals. (GNW, 1980:2-3)
They prefer the term FSR for research which includes the active participation
of the farmer. They also state: "Research on a sub-system can be considered
part of the FSR process if the connections with other sub-systems are recog-
nized and accounted for." (GNW, 1980: 3)
lishment. For small or part-time farmers in developed countries, FSR may have
the same benefits it does in Third World countries.
What Does FSR Have to Offer?
From the standpoint of agricultural economics, FSR potentially contri-
butes to problem-solving by improving understanding of the farming system, and
by enabling more effective development of technology for raising farm produc-
tivity. Improved understanding results from: (1) description and analysis of
interdependencies among components of the farming system, and between the farm
household and its environment; (2) the conceptual perspective on goals, con-
straints, and processes brought from other disciplines represented on the FSR
team; and (3) insights gained from including the farmer's viewpoint.
Improved understanding has a disciplinary payoff in terms of more power-
ful theory and analytical methods for research on the farm household.2
Improved understanding also leads to better diagnosis of problems and con-
straints within the farming system, and hence to the development of more pro-
ductive agricultural technology via new varieties, input combinations, cul-
tivation practices, etc. Thus, there is a close tie between problem identifi-
cation and prescription of solutions.
A domestic U.S. application of the systems approach to problem-solving is
research on integrated pest management. Chemical pest control is replaced by
a combination of biological and chemical controls, changes in crop mix and
cultivation, and more careful monitoring of pest populations. This approach
relies on information from several disciplines, including entomology, soil and
plant science, agricultural economics, and agricultural engineering, as well
as from the farmer. In West Africa, research on animal traction also
2By "more powerful" I mean having greater scope and/or predictive accuracy.
represents a systems approach to problem-solving. Low productivity and dec-
lining soil fertility under hand hoe bush fallow farming is addressed by a
mixed animal/crop farming system in which animal power breaks labor
bottlenecks, allows the incorporation of manure and crop residues to improve
soil fertility and thus crop yields, and provides a source of non-farm revenue
from animal-drawn carting.
The international brand of FSR was developed largely to address the prob-
lem of lack of adoption of improved agricultural technology. Low adoption
rates were a sign that important factors had been excluded in the technology
design process. FSR was intended to account for these missing factors, such
as: (1) interactions involving crop and animal enterprises and farm and non-
farm activities; (2) the performance of the technology under actual on-farm
conditions; and (3) economic and socio-cultural factors affecting acceptabil-
ity. Whether FSR-so defined--will in fact successfully overcome the "adop-
tion problem" remains to be seen.
Principles and Concepts of Agricultural Economics
re: FSR-Background for the Non-Economist
In studying a farming system, what analytical structure would be employed
by an agricultural economist? What variables and relationships would be exam-
ined? The following is a cursory and personally selective discussion of these
Farm households are considered to engage in several categories of
economic activity: production, consumption, marketing (buying and selling of
goods and services), and saving and investment. A common reductionistt"
approach is to study each of these activity categories in isolation from the
others. However, recent theorizing (both for U.S. and Third World farm types)
has emphasized the joint nature of these activities and the decisions
involved.3 For example, what the farm household produces is determined in part
by what it needs for consumption. Such interdependencies are particularly
salient for semi-subsistence farm households. This implies the need for a
more holistic approach, which has led to development of the "theory of the
A thumbnail sketch of the theory of the farm household includes the fol-
1. Households are assumed to maximize utility subject to various con-
straints. Utility is derived from household-produced farm and non-
farm goods, goods purchased from the market, and leisure. Goods and
leisure generate utility when they are consumed.
2. The constraints on utility maximization include:
(a) household production functions for farm and non-farm goods;
(b) a time availability constraint (time is used as an input to
household production, as well as for leisure);
(c) a budget constraint, which states that total expenditures must
be no greater than total income (including wage earnings and
income from assets).
As one moves from pure theory to applied research and farm management
studies, the variables and functional relationships associated with the farm
household must be specified in more concrete detail. Information of the fol-
lowing type is commonly sought:
1. Resource inventories. Economics is concerned fundamentally with the
allocation of resources to achieve specified goals. Thus, a start-
ing point is the identification of the stocks of resources held by
3The next section of the paper contains a more detailed discussion of these
the household, including land, buildings and machinery, working cap-
ital, family labor, and crop and livestock holdings.
2. Resource utilization. What are the flows of resources through the
farm system? Such flows include labor use, cash flow, machinery
use, etc. A high proportion of the data collected by large-scale
farm management surveys in the Third World results from documenting
these resource flows. Since household activities compete for a com-
mon set of resources, it is important to determine the flows of
goods and services among components within the farm household sys-
3. Description of household activities.
a. Farm production:
crop and livestock enterprises (rice, dairying, etc.)
-- production operations (weeding, harvesting, etc.)
b. Non-agricultural activities, including domestic household
tasks, crafts, trading, and other self-employed occupations
undertaken by household members.
c. Other off-farm activities, primarily wage employment.
4. Functional relationships and technical parameters.
a. Technical input-output coefficients for the production
b. The parameters of household consumption demand.
c. Marketing relationships, e.g., the timing and characteristics
of items purchased and sold.
d. Saving and investment behavior.
5. Costs and returns. These are a function of prices, which are a key
a. Operating costs associated with production and marketing.
b. Fixed costs, e.g., depreciation, interest, taxes, insurance,
etc. These are the costs of owning capital assets.
c. Sale prices for farm output; purchase prices for farm inputs
and family consumption items.
d. Wage rates for labor
6. Goals, attitudes and operating procedures.
a. Household goals, i.e., what does the household wish to achieve
through the use of its resources? Are its goals primarily pro-
fit or growth oriented, or more concerned with security and
maintenance of family well-being?
b. Preferences, including preferences for one type of farm
activity (e.g., cropping) over another (e.g., livestock rais-
ing). Attitudes toward risk are also important.
c. Operating procedures. Does the farmer have standard methods or
strategies which he follows in making decisions, e.g., stra-
tegies for reducing the impact of uncertainty in the production
7. Institutional and environmental variables.
a. How do the markets for resources function: for farm inputs and
outputs, for land and labor, and for credit? Does the farmer
have access to these markets? How reliable are they?
b. What is the physical environment of the farming system: soils,
rainfall, altitude, etc.?
c. Related to (a), are there important rules affecting access to
resources, such as rules governing land allocation and land
Household goals and preferences, and institutional and socio-cultural
variables, are a special focus of social scientists, including agricultural
economists. Consequently, one of the appropriate contributions of the agri-
cultural economist as an FSR team member is to view the farming system within
the larger institutional and policy framework, and to evaluate the impact of
market and price factors on the farm household's welfare.
Implementing FSR: Some Disciplinary Challenges
In previous discussions of the problems of organizing and implementing an
FSR program, the focus has been on the difficulties of a multidisciplinary
team approach, and on procedures for on-farm research (Norman, 1980; Rohrbach,
1980; CGIAR/TAC, 1978). Relatively little attention has been paid to ways in
which the analytical state of the art in the various disciplines poses obsta-
cles to successful research on farming systems.
In this section of the paper, I will try to sketch a few of the principal
theoretical, methodological, and empirical data limitations on the economic
analysis of the farm household system. My emphasis here is on the broad study
of existing farming systems and their response to proposed new technology or
policies, rather than on FSR for experimental development and field testing of
As in any systems approach, FSR does not require exhaustive enumeration
of system activities, but entails instead a focus on the essential ones. The
premise of FSR is that the study of farm households will be strengthened if
the scope of analysis is broadened to incorporate formerly neglected activi-
ties and interactions which are now recognized as crucial for understanding
This implies an increase in analytical complexity along several lines.
First, recognition that rural households are diverse rather than homogeneous,
- 10 -
and that variability in agroclimatic environment leads to location-specific
constraints and opportunities, calls for analysis of many rather than a few
household types. Second, the holistic perspective of FSR calls for an
increase in the number of household activities considered in the analysis.
Livestock activities are important and should be viewed in conjunction with
cropping activities (McDowell and Hildebrand, 1980). It may often be desir-
able to examine self-employed non-agricultural enterprises in addition to on-
farm production and off-farm wage employment. Related to this is a require-
ment for expansion of the set of variables which enter the analysis. This
stems both from the larger number of activities being studied, as well as from
the recognition that agroclimatic and socio-cultural factors have economic
Third, the analysis of farm household decisions must reflect reality more
closely. Aspects of importance are: (1) multiple goals and sequential
decision-making; (2) intrahousehold patterns of resource allocation; (3) the
interdependence between production decisions and those concerning credit,
marketing, consumption, savings, and investment; and (4) a dynamic, long-term
decision framework incorporating uncertainty.
Fourth, there is a need to recognize the interaction of the household
with its surrounding social and institutional environment. The analysis
should consider macro factors (market processes, institutions, and policies)
which impinge on the farm.
Farming systems studies are therefore threatened with the "curse of
dimensionality"-so many factors to analyze that a solution to the research
problem may be infeasible (Anderson and Hardaker, 1979). Agricultural econom-
ists must not only learn to effectively integrate expertise from specialized
fields within their own discipline (e.g., production, consumption, investment,
- 11 -
marketing, decision-making under uncertainty), but also from biological and
physical sciences, and other social sciences. This is an intellectually
demanding process, even for systems scientists who are at home with a holistic
rather than reductionist focus (Dillon, 1976).
Can this be accomplished within current theory and methodology in agri-
cultural economics? This question is clearly beyond the scope of a short
paper, but let us briefly examine the current disciplinary capacity to: (a)
incorporate household activities and interactions more comprehensively; and
(b) incorporate more realistic decision-making processes.
Expanding the scope of analysis is constrained partly by the available
theory. Neoclassical microeconomics has traditionally examined the
household's producer and consumer behavior separately. Beginning at least
with Mellor (1965), efforts have been made to integrate production and con-
sumption decisions in a household framework.4 Nakajima (1969) developed essen-
tially neoclassical models for subsistence and commercial family farms which
incorporated consumption and labor market participation, but not non-
agricultural production or (explicitly) leisure. Becker (1965) introduced the
concept of the household as a producing not just a consuming unit, with domes-
tic commodities produced with factors including time. Many recent economic
models of the agricultural household, e.g., Barnum and Squire (1979), draw on
Becker's work. The interdependence of production and consumption decisions
within the Barnum and Squire model is accomplished technically by specifying
consumption elasticities which contain terms for the effect of exogenous vari-
ables on farm profits, and hence on household income and consumption.
4Chayanov's much earlier model (1966, written in the early 1900's) of the
peasant household was holistic, but his definition of the peasant household as
relying entirely on family labor has restricted relevance today.
- 12 -
The advantages of such models are: (1) they examine the household in an
integrated framework which is consistent with the accepted theory of producer
and consumer maximizing behavior; (2) their format and assumptions are by now
fairly standard, hence easily interpreted by other researchers; (3) when
expressed in mathematical form their properties (e.g., elasticities) can be
Several offsetting disadvantages should be noted. First, identifying the
utility maximizing solution to such models and their corresponding elastici-
ties often requires the imposition of so many simplifying assumptions that
only very un-complex household types can be analyzed. For example, although
the model of Barnum and Squire includes only one variable input (labor), and
two outputs (rice and domestic goods) which are consumed along with leisure
and a purchased good, they found it necessary to limit the complexity of the
problem even further ". . by omitting certain decision variables in order to
arrive at a solution that has policy content."5 (Barnum and Squire, 1979:
A second, related disadvantage is the frequent need to restrict the func-
tional form of theoretical models in order to ensure the existence of
mathematical solutions. The result may conflict drastically with theoretical
axioms or observed reality. Third, although complexities such as multiple
goals, sequential decision-making, behavior over time, and stochastic varia-
bility have each been explored extensively in the literature (Day, 1977; Day
and Sparling, 1977), to this author's knowledge they have not yet been incor-
porated jointly in theoretical models of the farm household. Fourth, except
An extension of this model (Ahn, Singh, and Squire, 1979) incorporated pro-
duction of several crops using several inputs; farm commodities and profits
generated from a linear production system are fed into a linear expenditure
system to determine household consumption.
- 13 -
in linear models, the assumption of joint production or multiple inputs and
outputs tends to make solutions extremely tedious to obtain, or indeterminate
(Pollak and Wachter, 1975; Hart, 1978). Lastly, models of the household are
generally employed in the short run comparative statics context in which it is
possible to evaluate only small changes under ceteris paribus conditions.
Overall, theoretical models abstract from reality in a way which
emphasizes consistency, rigor, and computational convenience. This is more
appropriate in the context of disciplinary research than it may be in the con-
text of problem-solving research, which is the prime focus of FSR. Such
models are most useful in analyzing relatively simple farming systems (e.g.,
Asian rice monoculture)6 where restrictive assumptions and a short run per-
spective are justifiable.
Ability to expand the scope of farming systems studies is also limited by
available quantitative methods used to model the farm household. These
methods fall under three headings: econometrics, linear programming, and sys-
Econometric models are attractive in part because comparatively well-
accepted procedures are available for estimating and evaluating their struc-
tural parameters. Econometric models can potentially incorporate features
such as behavior over time and stochastic variability. For example, random
coefficients production models have been discussed by Swamy (1974), Mount
(1974), and Harville (1977), although apparently they have not yet been
applied to an integrated farm household model.
Linear programing (LP) has been a powerful, widely used tool. It is
flexible enough to incorporate features such as multiple inputs and outputs,
Such farming systems are simple only in the sense that the production process
can be approximated in terms of one output and a few inputs.
- 14 -
behavior over time, and the effect of uncertainty. One drawback of LP models
is their inherent tendency to give unrealistically one-sided optimal solu-
tions, e.g., over-specialized cropping patterns. Also, very careful scrutiny
is needed to establish whether an LP model is sound, or whether apparently
plausible results were "forced" by artificial, a-theoretical manipulation of
Systems simulation models offer even greater flexibility of form. Com-
plex features can be readily accommodated and solutions still obtained (John-
son and Rausser, 1977; Crawford, 1980a). Model specification can be eclectic
and behavioral, facilitating use in problem-solving research (Johnson, 1977).
The principal drawback is that simulation models often cannot be proven
theoretically consistent. A related problem is the inadequacy of standard
statistical procedures for evaluating how well a simulation model performs;
considerable subjective judgment is also required.
Other difficulties arise in modelling the link between events at the
individual farm level and aggregate effects at the macro level. FSR focuses
primarily on the farm level, but the literature generally recognizes the
importance of national policies and the regional agricultural economy as fac-
tors influencing the appropriate direction of new technology development.
However, both theoretical and quantitative models are limited in their ability
to predict the macro effects of introduced new technology or policy interven-
tions. At the formal level, there are problems of bias in aggregating the
results of individual farm models to obtain a picture of regional impact (Day,
1963); in general, it is not legitimate to assume that the whole is equal to
the sum of the parts. In addition, evaluating the impact of new technology or
This drawback is shared to some extent by econometric and systems simulation
- 15 -
policy interventions depends on the ability to analyze regional product and
factor markets. This moves the domain of the analysis from a partial to a
more general equilibrium framework, unless the situation can safely be simpli-
fied. Also, it is clear that the economic impact of a given development
intervention is heavily influenced by institutional and socio-cultural vari-
ables. However, rigorous prediction of socio-economic effects over time and
on a macro level is not yet feasible given available theory and analytical
methods in the social sciences.
Empirical Data Limitations
Achieving the holistic FSR approach will require better descriptive and
analytical understanding of several sub-systems of farm household activity
which hitherto have often been excluded. Perhaps the most important of these
is livestock activities, as noted above. Other areas deserving greater atten-
tion are: (1) self-employed non-agricultural occupations undertaken by farm
household members; (2) intrahousehold resource allocation and decision-making;
and (3) production and consumption/savings/investment behavior over time. All
four areas (including livestock) have been difficult to research for reasons
of required researcher mobility (in studying transhumant herders), access to
information (on self-employed and intrahousehold activities, often involving
women), and the cost and delay involved (time series data).
Current data collection methodologies are not entirely adequate, even for
partial analysis of farming systems. The recognized importance of considering
the diversity in farm household types and their agroclimatic environments--
rather than assuming homogeneity within broad categories (Crawford, 1980b;
Gerhart, 1975; Palmer-Jones, 1978) -- has encouraged reliance on detailed
large-scale farm management surveys. However, such surveys do not always
allow for the collection of the agronomic and socio-cultural information
- 16 -
needed for accurate identification of farm level constraints, or for the
analysis of complex features such as intercropped mixtures. Moreover, certain
kinds of information which may be crucial even for narrow economic analysis--
income, assets, food reserves--are known to be sensitive and hard to elicit
from respondents (Palmer-Jones, 1977).
As an alternative, the rapid survey technique used by some FSR practi-
tioners is less costly and easier to supervise. It brings knowledgeable
researchers into direct contact with farmers, using open-ended interviews in
place of minutely detailed questionnaires administered by enumerators. By
comparison to cost-route farm management surveys, from whose details the
essence of the farming system must be deduced, rapid surveys obtain less
detail but allow for the application of inductive reasoning and informed
intuition. Accordingly, they may ultimately give rise to a more profound
understanding of the farming system. A variant of this approach has been
developed by CIMMYT in East Africa (Collinson, 1979).
The Role of the Agricultural Economist in FSR
Although there is room for improvement in the systems research capacity
of the disciplinary tools of agricultural economics, agricultural economists
have a practical role to play in FSR programs at international and national
research centers, and as members of development project teams. As part of the
general goal of FSR, work by agricultural economists leads to improved under-
standing of existing farming systems. The particular contribution of agricul-
tural economists stems in part from the level at which they study the farming
system, namely the level of the whole farm or household rather than the level
of the plant/soil/water complex. Moreover, while assessing costs and returns
and resource constraints at the farm level is useful in solving some farmer
problems, economic analysis can and should define the farming system's
- 17 -
boundary more broadly, so that linkages with market institutions and the
effects of input/output price relationships are also examined. Because these
factors are critically important in determining the appropriate characteris-
tics of new technology, the FSR team has the opportunity to obtain an
indispensable contribution from the agricultural economist.
In turn, a wide range of disciplines can potentially assist in the solu-
tion of economic problems by elucidating critical variables, functional rela-
tionships, and constraints. For example, in studying investment behavior (or
the disposition of unspent income), it is desirable to know how social and
cultural factors determine the availability of investment opportunities and
who is allowed to take advantage of them. Or, in formulating a set of
improved cropping plans for farmer adoption, it is desirable to know which
crops are suited to local soils and rainfall, what their growing period is,
how yields respond to different cultivation practices, etc. More generally,
the economic feasibility of new technology or policy interventions is inextri-
cably tied to questions of technical, institutional, and socio-cultural feasi-
From a methodological standpoint, several disciplines are helpful in
facilitating economic research. As a mundane example, in mounting a farm sur-
vey, geographers can suggest which agroclimatic zones are relevant in strati-
fying the sample, statisticians can advise on sampling technique and data
analysis, biological scientists can advise on soil and water variables which
affect crop and livestock yields, and anthropologists can advise on important
cultural factors and techniques for informal interviewing.
- 18 -
Ahn, C. Y., Inderjit Singh, and Lyn Squire. 1979. "A Model of an Agricul-
tural Household in a Multi-Crop Economy: The Case of Korea." Invited
paper presented at the XVII Conference of The International Association
of Agricultural Economics, Banff, Canada, September 3-12.
Anderson, Jock R., and J. Brian Hardaker. 1979. "Economic Analysis in Design
of New Technologies for Small Farmers." Economics and the Design of
Small-Farmer Technology, Alberto Valdes, Grant M. Scobie, and John L.
Dillon (eds), pp. 11-26. Ames: Iowa State University Press.
Barnum, Howard N., and Lyn Squire. 1979. A Model of an Agricultural House-
hold: Theory and Evidence. World Bank Staff Occasional Papers No. 27.
Baltimore: John Hopkins University Press.
Becker, Gary S. 1965. "A Theory of the Allocation of Time." Economic Jour-
nal. 75: 493-517.
Chayanov, A. V. 1966. The Theory of the Peasant Economy. Homewood: Richard
Collinson, Michael. 1979. "Micro-Level Accomplishment and Challenges for the
Less Developed World." Paper presented at the XVII Conference of the
International Association of Agricultural Economics, Banff, Canada, Sep-
Consultative Group on International Agricultural Research, Technical Advisory
Committee (CGIAR/TAC). 1978. "Farming Systems Research at the Interna-
tional Agricultural Research Centers." Mimeo. September.
Crawford, Eric W. 1980a. "A Stochastic Farm System Simulation Model, with
Emphasis on Experimentation." Department of Agricultural Economics Staff
Paper No. 80-81, Michigan State University, East Lansing, Michigan.
1980b. "Understanding, Quantification, and Modelling in Farming
Systems Research: Results of a Simulation Study in Northern Nigeria."
Department of Agricultural Economics Staff Paper No. 80-19, Michigan
State University, East Lansing, Michigan.
Day, Richard H. 1977. "On Economic Optimization: A Nontechnical Survey."
A Survey of Agricultural Economics Literature, Vol. 2: Quanti-
tative Methods in Agricultural Economics, 1940 to 1970s, George G. Judge
et al. (eds), pp. 57-92. Minneapolis: University of Minnesota Press.
1963. "On Aggregating Linear Programming Models of Production."
Journal of Farm Economics. 45:797-813
Day, Richard H., and Edward Sparling. 1977. "Optimization Models in Agricul-
tural and Resource Economics." A Survey of Agricultural Economics
Literature, Vol. 2, Judge et al. (eds), pp. 93-127. Minneapolis:
University of Minnesota Press, 1977.
- 19 -
Dillon, John, L. 1976. "The Economics of Systems Research." Agri-
cultural Systems. 1: 5-22.
Gerhart, John. 1975. "The Diffusion of Hybrid Maize in Western Kenya--
Abridged by CIMMYT." Mexico City: CIMMYT.
Gilbert, E. H., D. W. Norman, and F. E. Winch. 1980. Farming Systems
Research: A Critical Appraisal. Rural Development Paper No. 6. East
Lansing, Department of Agricultural Economics, Michigan State University.
Hart, Gillian P. 1978. "Labor Allocation Strategies in Rural Javanese House-
holds." Ph.D. thesis, Cornell University.
Harville, D. A. 1977. "Maximum Likelihood Approaches to Variance Component
Estimation and to Related Problems." Journal of the American
Statistical Association. 72:320-338.
Hildebrand, P. E. 1977. "Generating Small Farm Technology: An Integrated
Multi-Disciplinary System." Paper presented at 12th West Indian Agricul-
tural Economics Conference, 24-30 April 1977, Antigua.
Johnson, Glenn L. 1977. "Contributions of Economists to a Rational
Decision-Making Process in the Field of Agricultural Policy." Decision-
Making and Agriculture, Theodore Dams and Kenneth E. Hunt (eds), pp. 25-
46. Oxford: Alden Press.
Johnson, S. R., and Gordon C. Rausser. 1977. "Systems Analysis and Simula-
tion: A Survey of Applications in Agricultural and Resource Economics."
A Survey of Agricultural Economics Literature, Vol. 2, Judge et al.
(eds), pp. 157-304. Minneapolis: University of Minnesota Press.
McDowell, R. E., and P. E. Hildebrand. 1980. "Integrated Crop and Animal
Production: Making the Most of Resources Available to Small Farms in
Developing Countries." New York: The Rockefeller Foundation, January,
Mellor, John W. 1965. "The Subsistence Farmer in Traditional Economies."
Paper presented at the ADC Seminar on Subsistence and Peasant Economies,
East-West Center, Honolulu, Hawaii, February-March.
Mount, T. D. 1974. "Applications of Variance Components in Economics." Cor-
nell University, Department of Agricultural Economics Staff Paper 74/7.
Nakajima, Chihiro. 1969. "Subsistence and Commercial Family Farms: Some
Theoretical Models of Subjective Equilibrium." Subsistence Agriculture
and Economic Development, Clifton R. Wharton, Jr. (ed), pp. 165-185.
Chicago: Aldine Publishing.
Norman, David W. 1980. "The Farming Systems Approach: Relevancy for the
Small Farmer." Rural Development Paper No. 5, Department of Agricultural
Economics, Michigan State University.
- 20 -
Palmer-Jones, R. W. 1978. "Peasant Differentiation in Rural Hausaland."
Mimeo. Department of Agricultural Economics and Rural Sociology, Insti-
tute for Agricultural Research, Zaria, Nigeria.
1977. "Field Research Methods Used in Farm Management Studies."
Mimeo. Department of Agricultural Economics and Rural Sociology, Insti-
tute for Agricultural Research, Zaria, Nigeria.
Pollak, R. A., and M. L. Wachter. 1975. "The Relevance of the Household Pro-
duction Function and Its Implications for the Allocation of Time."
Journal of Political Economy. 83:255-278.
Rohrbach, David D. 1980. "A Discussion of Issues Relevant to the Development
and Implementation of a Farming Systems Research Program." Mimeo. Wash-
ington, D.C.: USDA/OICD.
Swamy, P.A.V.B. 1974. "Linear Models with Random Coefficients." Frontiers
of Econometrics, P. Zarembka (ed). New York: Academic Press.